Official code for "F5R-TTS: Improving Flow-Matching based Text-to-Speech with Group Relative Policy Optimization"

Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation

本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）

Qwen3-TTS is an open-source series of TTS models developed by the Qwen team at Alibaba Cloud, supporting stable, expressive, and streaming speech generation, free-form voice design, and vivid voice…

MOSS-Audio-Tokenizer is a Causal Transformer-based audio tokenizer built on the CAT architecture. Trained on 3M hours of diverse audio, it supports streaming and variable bitrates, delivering SOTA …

Train the next generation of TTS systems.

Fast and memory-efficient exact attention

A powerful 3B-parameter, LLM-based Reinforcement Learning audio edit model excels at editing emotion, speaking style, and paralinguistics, and features robust zero-shot text-to-speech

Open-Source Frontier Voice AI

The best ChatGPT that $100 can buy.

1 min voice data can also be used to train a good TTS model! (few shot voice cloning)

pyright fork with various type checking improvements, improved vscode support and pylance features built into the language server

12306接口抢票

Edit, preview and share mermaid charts/diagrams. New implementation of the live editor.

An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

Awesome speech/audio LLMs, representation learning, and codec models

Audio Large Language Models

Your one-stop solution for voice dataset creation

Towards Human-Sounding Speech

SOTA Open Source TTS

Reverse Engineering of Supervised Semantic Speech Tokenizer (S3Tokenizer) proposed in CosyVoice

Text Normalization & Inverse Text Normalization

基于SparkTTS、OrpheusTTS等模型，提供高质量中文语音合成与声音克隆服务。

[ACL 2024] Official PyTorch code for extracting features and training downstream models with emotion2vec: Self-Supervised Pre-Training for Speech Emotion Representation

Official code for "EmoVoice: LLM-based Emotional Text-To-Speech Model with Freestyle Text Prompting"

How to use our public wav2vec2 dimensional emotion model

This is an evolving repo for the paper "Towards Controllable Speech Synthesis in the Era of Large Language Models: A Systematic Survey".

Added vLLM support to IndexTTS for faster inference.

Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.